prolactin-induced pak1 tyrosyl phosphorylation promotes ...[17], while expression of kinase dead pak...

RESEARCH ARTICLE Open Access

Prolactin-induced PAK1 tyrosylphosphorylation promotes FAKdephosphorylation, breast cancer cellmotility, invasion and metastasisAlan Hammer and Maria Diakonova*

Abstract

Background: The serine/threonine kinase PAK1 is an important regulator of cell motility. Both PAK1 and thehormone/cytokine prolactin (PRL) have been implicated in breast cancer cell motility, however, the exactmechanisms guiding PRL/PAK1 signaling in breast cancer cells have not been fully elucidated. Our lab haspreviously demonstrated that PRL-activated tyrosine kinase JAK2 phosphorylates PAK1 on tyrosines 153, 201, and285, and that tyrosyl phosphorylated PAK1 (pTyr-PAK1) augments migration and invasion of breast cancer cells.

Results: Here we further investigate the mechanisms by which pTyr-PAK1 enhances breast cancer cell motility inresponse to PRL. We demonstrate a distinct reduction in PRL-induced FAK auto-phosphorylation in T47D andTMX2-28 breast cancer cells overexpressing wild-type PAK1 (PAK1 WT) when compared to cells overexpressingeither GFP or phospho-tyrosine-deficient mutant PAK1 (PAK1 Y3F). Furthermore, pTyr-PAK1 phosphorylates MEK1 onSer298 resulting in subsequent ERK1/2 activation. PRL-induced FAK auto-phosphorylation is rescued in PAK1 WTcells by inhibiting tyrosine phosphatases and tyrosine phosphatase inhibition abrogates cell motility and invasion inresponse to PRL. siRNA-mediated knockdown of the tyrosine phosphatase PTP-PEST rescues FAK auto-phosphorylation in PAK1 WT cells and reduces both cell motility and invasion. Finally, we provide evidence thatPRL-induced pTyr-PAK1 stimulates tumor cell metastasis in vivo.

Conclusion: These data provide insight into the mechanisms guiding PRL-mediated breast cancer cell motility andinvasion and highlight a significant role for pTyr-PAK1 in breast cancer metastasis.

Keywords: PAK1, FAK, Prolactin, Tyrosyl phosphorylation, Breast cancer cells

BackgroundProlactin (PRL) is a peptide hormone/cytokine that istypically secreted from the anterior pituitary gland, andhas been found to be locally produced in various otherorgans such as the prostate, uterus, and mammary gland(for review [1]). Upon PRL binding, PRL-receptor (PRLR)dimerizes resulting in activation of the non-receptortyrosine kinase JAK2 (Janus kinase 2) and subsequentdownstream signaling cascades including signal trandu-cers and activators of transcription (STATs), mitogen acti-vated protein kinases (MAPKs), including ERK1/2, and

phosphoinositol-3 kinase pathways (for review [2]). PRLsignaling at both an endocrine and paracrine/autocrinelevels regulates a variety of physiological processes in aneclectic range of tissues (for review [3]). There is mount-ing evidence that PRL plays a significant role in breastcancer. The PRLR has been found in the vast majority ofhuman breast cancers and PRL signaling has been impli-cated in breast cancer cell proliferation, survival, motilityand angiogenesis (for review [2]). Furthermore, elevatedcirculating PRL levels have been positively correlated withbreast cancer metastasis and PRLR-deficient mice haveprevention of neoplasia progression into invasive carcin-oma [4–7]. Importantly, PRL has been noted as a chemo-attractant for breast cancer cells and augments tumor

* Correspondence: [email protected] of Biological Sciences, University of Toledo, 2801 W. BancroftStreet, Toledo 43606-3390, OH, USA

© 2016 The Author(s). Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.

Hammer and Diakonova BMC Cell Biology (2016) 17:31 DOI 10.1186/s12860-016-0109-5

http://crossmark.crossref.org/dialog/?doi=10.1186/s12860-016-0109-5&domain=pdfhttp://orcid.org/0000-0002-4554-2730mailto:[email protected]://creativecommons.org/licenses/by/4.0/http://creativecommons.org/publicdomain/zero/1.0/

metastasis in nude mice [8, 9]. However, the exact mecha-nisms guiding PRL-induced cell migration and tumor me-tastasis are not fully understood.We have implicated the serine/threonine kinase PAK1

(p21-activated kinase-1) as a substrate of PRL-activatedJAK2 [10]. PAK1 has been associated with breast cancerprogression (for review [11]). Aberrant expression/acti-vation of PAK1 has been described in breast cancer aswell as among several other cancers including brain,pancreas, colon, bladder, ovarian, hepatocellular, urinarytract, renal cell carcinoma, and thyroid cancers (for review[12]). The PAK1 gene lies within the 11q13 region and11q13.5→ 11q14 amplifications involving the PAK1 locusare present in 17 % of breast cancers [13, 14]. PAK1 over-expression was observed in over half of observed breasttumor specimens [15] and PAK1 expression is correlatedwith tumor grade [16–18]. In transgenic mouse models,hyperactivation of PAK1 promotes mammary gland tumorformation [19]. Interestingly, overexpression of constitu-tively active PAK1 T423E in non-invasive breast cancercells stimulates cell motility and anchorage independence[17], while expression of kinase dead PAK in highly inva-sive breast cancer cells significantly reduces cell invasive-ness [20]. PAK1 kinase activity promotes directional cellmotility and is a major regulator of the actin cytoskeleton(for review [11]). We have previously demonstrated thatPRL-activated JAK2 directly phosphorylates PAK1 on ty-rosines 153, 201, and 285 [10]. We have also demon-strated that tyrosyl phosphorylated PAK1 (pTyr-PAK1)enhances PRL-mediated cell invasion via MAPK activationand increased matrix metalloproteinase expression [21] aswell as cell motility through increased phosphorylation ofactin-crosslinking protein filamin A ([22]; reviewed in[23]). Additionally, PRL-induced pTyr-PAK1 is localizedat small adhesion complexes at the cell periphery and reg-ulates adhesion turnover in breast cancer cells, a processthat is absolutely critical for cell motility [24].Cell motility is essential in the regulation of many sig-

nificant biological processes including embryogenesis,wound healing, and immune responses; however aberrantcell migration is present in malignant cancers and resultsin the establishment of tumors in distant tissues. Cell mo-tility is a highly coordinated process that requires tightregulation of the actin cytoskeleton, cell-matrix adhesionturnover, and complex intracellular signaling cascades.The tyrosine kinase focal adhesion kinase (FAK) has beenimplicated as an important regulator of cell motility (forreview [25]). FAK is localized to cell/matrix adhesions andis activated by integrin engagement to the extracellularmatrix as well as by several other extracellular ligands (forreview [26]). Auto-phosphorylation of FAK at tyrosine 397(Y397) promotes FAK activation and recruits SH2- andSH3-domain containing proteins, most notably c-Src,leading to Src-mediated FAK activation and activation of

Src/FAK signaling pathways, including the ERK MAPKsignaling cascade (for review [26]). FAK activation hasbeen most well implicated in the positive regulation of cellmotility (for review [26, 27]). However, recently more evi-dence has demonstrated a controversial role for FAK as anegative regulator of cancer cell migration [28–30].Here we extend our knowledge on the role for

pTyr-PAK1 in PRL-induced breast cancer cell motilityand invasion. We use T47D and TMX2-28 breast cancercells stably overexpressing GFP, PAK1 WT, or tyrosylphosphorylation-deficient mutant of PAK1 in which thethree JAK2 phosphorylation sites have been mutated tophenylalanine (PAK1 Y3F). These cells were previouslycharacterized in [22, 24] and [21]. We demonstrate herethat tyrosyl phosphorylation of PAK1 in response to PRLregulates PTP-PEST-dependent FAK dephosphorylation,resulting in augmented breast cancer cell migration andinvasion and proposed the mechanism explaining thesefindings. Furthermore, we provide in vivo evidence thatPRL-induced pTyr-PAK1 increases breast cancer cellmetastasis. Taken together, these data suggest thatPRL-mediated pTyr-PAK1 is important in regulatingthe dynamic activation of FAK and subsequent breastcancer cell migration and invasion.

MethodsAntibodies and reagentsPolyclonal αpY397-FAK (Abcam), monoclonal αFAK(EMD Millipore), polyclonal αpS298-MEK (Cell Signaling),monoclonal αMEK (GeneTex), monoclonal αphospho-ERK1/2 (pT202/Y204) and polyclonal αERK1/2 (CellSignaling), monoclonal αmyc (9E10, Santa Cruz Biotech-nology), and αγ-tubulin (Sigma-Aldrich) were used forimmunoblotting. Na3VO4 was purchased from Sigma.siRNA and primers for PTP-PEST were purchased fromSanta Cruz Biotechnology. Control nontargeting siRNAwas purchased from Cell Signaling. Human PRL was pur-chased from the National Hormone and Peptide Program(Dr. Parlow, National Institute of Diabetes and Digestiveand Kidney Diseases).

Cell cultureProlactin receptor- and estrogen receptor-positive T47Dcells stably overexpressing GFP, myc-tagged PAK1 WT,and myc-tagged PAK1 Y3F were described previously[22, 24]. T47D clones were maintained in RPMI 1640medium (Corning Cellgro, Corning, Inc) supplementedwith 10 % fetal bovine serum (FBS; Sigma-Aldrich) andinsulin (Sigma-Aldrich). Prolactin receptor-positive butestrogen receptor-negative TMX2-28 cells (a variant ofthe MCF-7 breast cancer cell line [31]) and their clonesstably overexpressing GFP, PAK1 WT or PAK1 Y3F weredescribed previously [21] and maintained in DMEMsupplemented with 10 % fetal bovine serum. The levels

Hammer and Diakonova BMC Cell Biology (2016) 17:31 Page 2 of 13

of overexpressed PAK1 WT and PAK1 Y3F were roughlyestimated to be around 20-fold over the level of endogen-ous PAK1 in both T47D cells and TMX2-28 cells. MCF-7cells were kindly donated by Dr. Ethier (University ofMichigan) and T47D cells were purchased from theATCC. TMX2-28 cells were kindly donated by Dr. Eisen-mann (University of Toledo, OH).

Assessing FAK, MEK, and ERK phosphorylationT47D or TMX2-28 clones were seeded into 6-welldishes and deprived of serum for 72 h before treatmentwith or without PRL (200 ng/ml) for the indicated times.Cells were lysed and proteins were resolved by SDS-PAGE followed by immunoblotting with the indicatedantibodies. Fold FAK, MEK, and ERK activation wasassessed by densitometric analysis of αphospho-proteinbands normalized to αtotal-protein bands using ImageJsoftware. To assess FAK activation in T47D clones inthe absence of tyrosine phosphatase activity, cells weretreated with 100 ng/ml of Na3VO4 for one hour beforetreatment with or without PRL (200 ng/ml) for the indi-cated times. Cells were lysed and proteins were resolvedby SDS-PAGE followed by immunoblotting with the in-dicated antibodies. FAK activation was assessed bydensitometric analysis of αpY397-FAK bands normalizedto αFAK bands using ImageJ software.

PTP-PEST knockdownPTP-PEST siRNA or control nontargeting siRNA weretransfected into T47D or TMX2-28 cells using Lipofecta-mine RNAiMAX (Invitrogen) according to the manufac-turer’s instructions. The final concentration of the siRNAwas 100 nM. Knockdown of PTP-PEST mRNA wasassessed by RT-PCR method using PTP-PEST primers.To assess PRL-induced FAK activation in the absence

of PTP-PEST, T47D and TMX2-28 clones were transfectedwith PTP-PEST siRNA, deprived of serum for 48 h, andtreated with or without PRL for the indicated times. Cellswere lysed and proteins were resolved by SDS-PAGEfollowed by immunoblotting with the indicated antibodies.

Cell viabilityTo assess cell viability in the presence of 100 ng/mlNa3VO4 for 48 h, equal numbers of T47D cells were re-suspended in deprivation media (RPMI 1640 mediumsupplemented with 1 % BSA) with or without PRL(200 ng/ml) and Na3VO4 (100 ng/ml) then seeded into a96-well plate. After 48 h, cells were subjected to theVybrant® MTT Cell Proliferation Assay (Molecular Probes)according to the manufacturer’s instructions.

Cell migration and cell invasion assaysCell migration and cell invasion assays were performedas we described previously [21, 22]. Equal cell numbers of

the T47D (1 × 106 cells/chamber) or TMX2-28 (0.5 × 106

cells/chamber) stable cell lines for each condition wereplaced in deprivation media with or without 100 ng/mlNa3VO4 in the upper chamber of a Boyden chamber(8.0 μm pores, Corning, Inc) (migration assay) or aBoyden chamber (8.0 μm pores), coated with Matrigel(BD Biosciences) (invasion assay). Deprivation media withor without 200 ng/ml PRL was placed in the lower cham-ber. Cells were allowed to migrate or invade for 48 h, afterwhich the cells remaining in the upper chamber were re-moved from the upper chamber by a cotton swab. Cellsfrom five separate fields that had migrated through thepores of the membrane to the underside of the filter werecounted after fixation with 4 % formalin (Sigma) andstaining with Differential Quik Stain (Polysciences, Inc).Brightfield images of migrated/invaded cells were acquiredon an inverted Olympus IX81 microscope using LUCPlanFLN 40× objective lens and wide field WHN 10X eyepiece(Olympus, Tokyo, Japan).To assess the effect of PTP-PEST knockdown on cell

migration and invasion, T47D and TMX2-28 stable cloneswere transfected with PTP-PEST siRNA. After 24 h,cells were placed in deprivation media in the upperchamber of a Boyden chamber (migration assay) or aBoyden chamber coated with Matrigel (invasion assay).Cells were allowed to migrate/invade for 48 h and proc-essed as described above.

In vivo metastasisTMX2-28 clones stably overexpressing GFP, myc-PAK1WT or myc-PAK1 Y3F were inoculated directly into mam-mary fat pad of NSG (NOD/SCID/ IL2Rgamma) femalemice. hPRL (20 μg/100 μl) was injected subcutaneouslyevery other day for 8 weeks and mice were terminated in12 weeks. 8 mice were used for TMX2-28 PAK1 clone, 6mice for TMX2-28 PAK1 Y3F clone and 4 mice forTMX2-28 GFP clone. Mouse experimental procedureswere performed in the animal research core of Lerner Re-search Institute, Cleveland Clinic (Dr. Lindner), and wereapproved by the Institutional Animal Care and Use Com-mittee, Cleveland Clinic. The first half of tumors and lungsfrom mice was frozen and kept at −80 °C. Before use, thetissues were homogenized in RIPA buffer with protease in-hibitors (50 mM Tris-HCl, 150 mM NaCl, 2 nM EGTA,1 % Triton X-100, aprotinin 10 μg/ml, leupeptin, 10 μg/ml,pH7.5; 500 μL per 10 mg tissue) at 4 °C. Homogenized tis-sues were rotated in RIPA buffer for 1 h at 4 °C to ensurecell lysis. Samples were centrifuged at 10,000 g to pelletdebris and protein concentration in supernatant was deter-mined by Bradford assay. Proteins were separated by SDS-PAGE and transferred to PVDF membrane. Lysates ofTMX2-28 PAK1 WT cells were loaded as a control forPAK1-myc position in the gels. Membranes were probedwith anti-myc to detect myc-PAK1 WT or Y3F in the


tissues and anti-tubulin for loading control. The secondhalf of tumors and lungs was fixed with 10 % formalin andembedded in paraffin. Immunohistochemistry usingparaffin-embedded sections was done as described previ-ously [24]. Briefly, formalin-fixed, paraffin-embedded sec-tions were boiled for 15 min in 0.01 M sodium citratebuffer (pH 6.0) to expose antigenic epitopes. Sections wereblocked with 2.5 % normal horse serum for 30 min andthen incubated overnight with anti-myc (1:100) or controlpre-immune serum. The biotinylated secondary antibodywas used followed by streptavidin horseradish peroxidasesolution (R.T.U. Vectstatin universal quick kit, Vector La-boratories). The chromogen was 3,3’ diaminobenzidine(ImmPACT DAB kit, Vector Laboratories). Staining withpre-immune serum was negligible (not shown).

Statistical analysisData from at least 3 separate experiments were pooledand analyzed using 1-way ANOVA plus Tukey’s honestsignificant difference test. Differences were considered tobe statistically significant at P < 0.05. Results are expressedas the mean ± SE.

ResultsTyrosyl phosphorylated PAK1 negatively regulates FAKauto-phosphorylationWe have previously demonstrated that PRL promotesbreast cancer cell motility in a pTyr-PAK1-dependentmanner [22]. In an attempt to understand the pTyr-PAK1-

dependent mechanism that regulates PRL-induced cellmotility, we first examined the auto-phosphorylation ofFAK in response to PRL, as FAK is an important regulatorof cell motility (for review [25]). T47D GFP (control),PAK1 WT, or PAK1 Y3F (phospho-tyrosine-deficient mu-tant) clones were treated with PRL over a time-course andwhole cell lysates (WCL) were analyzed for FAK auto-phosphorylation at Y397, which is critical for Src/FAKinteraction and maximal FAK activation (reviewed in [32]).PRL treatment led to maximal FAK auto-phosphorylationin 15 min in control GFP cells (Fig. 1a, left blot; Fig. 1b,solid line). On the contrary, there was no significantY397-FAK auto-phosphorylation in response to PRL inthe PAK1 WT cells (Fig. 1a, middle blot; Fig. 1b, dashedline), suggesting that PRL-induced pTyr-PAK1 has anegative effect on FAK auto-phosphorylation. FAK wasmaximally auto-phosphorylated by PRL in 7.5 min inPAK1 Y3F cells (Fig. 1a, right blot; Fig. 1b, dotted line).Similar results were obtained in TMX2-28 (estrogen-re-ceptor-negative sub-line of the MCF-7 breast cancer cells[31]) stably overexpressing GFP, myc-PAK1 WT, or myc-PAK1 Y3F (Fig. 2d, anti-pY397-FAK and anti-FAK blots)indicating that this finding was not restricted to T47Dcells. It is important to note that PRL-dependent Y395-FAK phosphorylation was transient in both T47D GFPand T47D PAK1 Y3F clones. Our data suggest that tyrosylphosphorylation of PAK1 in response to PRL promotesFAK dephosphorylation and tyrosines 153, 201 and 285 ofPAK1 are responsible for this effect.

Fig. 1 Tyrosyl phosphorylation of PAK1 negatively regulates PRL-induced FAK auto-phosphorylation. a Whole cell lysates (WCL) of T47D cellsstably overexpressing GFP, PAK1 WT, or PAK1 Y3F treated with PRL (200 ng/ml) for the indicated times were probed for FAK auto-phosphorylationusing αpY397-FAK antibody. The expression levels of γtubulin were used as an internal loading control. b Graph represents the densitometric analysisof the bands obtained for pY397-FAK normalized to total FAK for at least 3 independent experiments. The solid line represents T47D GFP cells, thedashed line represents T47D PAK1 WT cells, and the dotted line represents T47D PAK1 Y3F cells. Bars represent mean ± SE. *P < 0.05 compared withthe same cells not treated with PRL


Tyrosyl phosphorylation of PAK1 promotes S298-MEK1phosphorylation and ERK activation in response to PRLTo uncover the mechanism by which pTyr-PAK1 mayregulate FAK phosphorylation, we assessed S298-MEKphosphorylation and consequent ERK1/2 activation(dual phosphorylation of T202 and Y204 of ERK1/2 me-diates ERK activity [33, 34]) in response to PRL becausea PAK1/MEK/ERK signaling cascade has been impli-cated in Ras-mediated FAK dephosphorylation [30]. PRLpromoted PAK1-dependent MEK phosphorylation 6-foldin as early as 7.5 min and maximal 8-fold MEK phos-phorylation after 15 min in T47D PAK1 WT cells(Fig. 2a, middle blot, Fig. 2b). PRL also induced pS298-

MEK signal in the T47D GFP and T47D PAK1 Y3F cellsalbeit slower and to a lesser extent when compared tothe PAK1 WT cells (Fig. 2a, left and right blots, Fig. 2b).Subsequently, ERK1/2 was phosphorylated in responseto PRL in all three T47D clones, however earlier and toa much greater extent in the PAK1 WT cells when com-pared to GFP and PAK1 Y3F cells (Fig. 2a and c). Similarresults were obtained in TMX2-28 GFP, PAK1 WT andPAK1 Y3F cell clones (Fig. 2d, anti-pS298-MEK, anti-p-ERK1/2, anti-MEK and anti-ERK1/2 blots). These datasuggest that PAK1 tyrosyl phosphorylation promotesPAK-dependent MEK phosphorylation and ERK activa-tion in response to PRL.

Fig. 2 Tyrosyl phosphorylation of PAK1 promotes S298-MEK1 phosphorylation and ERK activation in response to PRL. a WCL of T47D cells stablyoverexpressing GFP, PAK1 WT, or PAK1 Y3F treated with PRL (200 ng/ml) for the indicated times were probed for MEK phosphorylation usingαpS298-MEK and ERK1/2 activation using αphospho-ERK1/2 (pT202/Y204) antibodies. b, c Graphs represent the densitometric analysis of thebands obtained for phospho-MEK (b) or phospho-ERK1/2 (c) normalized to total MEK or ERK1/2, respectively, for at least 3 independentexperiments. Bars represent mean ± SE . *P < 0.05 compared with cells expressing GFP with the same treatment. d WCL of TMX2-28 cellsstably overexpressing GFP, PAK1 WT, or PAK1 Y3F treated with PRL (200 ng/ml) for the indicated times were probed with the indicated antibodies. Theexpression levels of γtubulin were used as an internal loading control


Protein tyrosine phosphatase inhibition rescuesPRL-mediated auto-phosphorylation of FAKIn order to determine whether tyrosine phosphatases areinvolved in the negative effect of pTyr-PAK1 on FAKauto-phosphorylation, we assessed Y397- FAK phos-phorylation in response to PRL in the presence or ab-sence of Na3VO4, a tyrosine phosphatase inhibitor.T47D PAK1 WT cells were treated with vehicle orNa3VO4 for one hour before PRL treatment and WCLwere assessed for pY397-FAK (Fig. 3). As expected,there was no PRL-mediated increase in FAK tyrosylphosphorylation in vehicle treated cells (Fig. 3a, lanes 1and 2). However, phosphatase inhibition led to a signifi-cant increase in both basal and PRL-induced FAK auto-phosphorylation (Fig. 3a, lanes 3 vs. 1 and 4 vs. 2). Toconfirm that phosphatase activity is important for thepTyr-PAK1-dependent effect of PRL on FAK dephos-phorylation, all three T47D clones were subjected to aPRL time-course in the presence of Na3VO4 and WCLwere assessed for Y397 FAK auto-phosphorylation(Fig. 3b, c). In the presence of Na3VO4, PRL treatmentactivated FAK in all three cell lines regardless of thestatus of PAK1 tyrosyl phosphorylation (Fig. 3b, c). Fur-thermore, FAK remained phosphorylated until the end

of the PRL time-course in all three clones in the pres-ence of Na3VO4.Next we aimed to determine whether tyrosine phos-

phatase PTP-PEST, which dephosphorylates FAK at Y397[30], participates in PRL- and PAK1-dependent lack ofFAK auto-phosphorylation. PTP-PEST silencing in T47Dand TMX2-28 clones was confirmed by RT-PCR method(Fig. 4a). We performed siRNA-based silencing of PTP-PEST in T47D (Fig. 4b) and TMX2-28 (Fig. 4c) clones,treated the cells with or without PRL and assessed forY397 FAK auto-phosphorylation. Indeed, PTP-PEST silen-cing rescued Y397-FAK phosphorylation in PAK1 WTcells to similar levels to that of GFP and PAK1 Y3F cells inresponse to PRL (Fig. 4b, c). On the contrary, there was nosignificant Y397-FAK auto-phosphorylation in response toPRL in the PAK1 WT clones transfected with controlsiRNA (Fig. 4b, c).These data suggest that tyrosine phosphatase activ-

ity of PTP-PEST is responsible for the apparent lackof FAK auto-phosphorylation in response to PRL inPAK1 WT cells. Given the complexity of these sig-naling cascades, it is likely that additional signalingmolecules are also involved in the modulation of FAKphosphorylation.

Fig. 3 Protein tyrosine phosphatase inhibition rescues PRL-mediated FAK auto-phosphorylation in T47D WT cells. a Tyrosine phosphatase inhibition byNa3VO4 permits PRL-induced FAK auto-phosphorylation in PAK1 WT cells. WCL of T47D PAK1 WT cells treated with either vehicle (veh) orNa3VO4 (100 ng/ml) for 1 h before PRL (200 ng/ml) treatment were probed for FAK auto-phosphorylation by αpY397-FAK antibody. b FAKis auto-phosphorylation in T47D GFP, PAK1 WT, and PAK1 Y3F cells in response to PRL in the presence of Na3VO4. The cells were treatedwith Na3VO4 as in A and with PRL (200 ng/ml) for the indicated times. FAK auto-phosphorylation was assessed as in A. The expression levels of γtubulinwere used as an internal loading control. c Graph represents the densitometric analysis of the bands obtained for pY397-FAK normalized to total FAKfor at least 3 independent experiments. Bars represent mean ± SE. *P < 0.05 compared with the same cells not treated with PRL


Protein tyrosine phosphatase inhibition impedes PRL-mediated T47D and TMX2-28 cell migration and invasionTo investigate whether tyrosine phosphatases regulatePRL/pTyr-PAK1-dependent T47D breast cancer cell mi-gration, we examined migration of T47D clones in thepresence and absence of PRL and Na3VO4 for 48 h. Asdynamic tyrosyl phosphorylation events are crucial tomany cellular processes, it was important to test whetherphosphatase inhibition for an extended period of 48 hhad any cytotoxic effect. The cell viability was assessedin serum deprived T47D cells treated with or withoutPRL and Na3VO4 (Fig. 5c). Na3VO4 had no significantcytotoxic effect on any of the three stable cell lines in

the presence or absence of PRL (Fig. 5c). Next, the effectof tyrosine phosphatase inhibition on PRL-mediated cellmigration was assessed using a transwell migration assay.Equal numbers of T47D GFP, PAK1 WT and PAK1 Y3Fcells were seeded into the upper part of a Boyden chamberwith or without Na3VO4 and PRL or vehicle were addedto the bottom part. The number cells that migratedthrough the chamber towards PRL were counted (Figs. 5aand d). As we demonstrated previously [22], PRL stimu-lated cell migration to a greater extent in PAK1 WT cellswhen compared to GFP and PAK1 Y3F cells in the ab-sence of Na3VO4 (Fig. 5d, veh). However, phosphatase in-hibition by Na3VO4 completely abolished cell migration inresponse to PRL in all T47D clones (Fig. 5d, Na3VO4).These data suggest that phosphatase activity is requiredfor pTyr-PAK1-induced cell migration.Cell migration is a key step in cell invasion so we de-

cided to assess the effect of phosphatase inhibition oncell invasion. Equal numbers of T47D GFP, PAK1 WTand PAK1 Y3F cells were seeded into the upper chamberof a Boyden chamber coated with Matrigel, in the pres-ence of either Na3VO4 or vehicle. Deprivation media withor without PRL (200 ng/ml) was added to the lower cham-ber of the Boyden chamber. The number of cells thatinvaded through the Matrigel towards PRL was counted.As we demonstrated previously [21], PRL stimulated cellinvasion to a greater extent in PAK1 WT cells when com-pared to GFP and PAK1 Y3F cells in the absence ofNa3VO4 (Fig. 5b and e, veh). However, Na3VO4-mediatedtyrosine phosphatase inhibition abolished cell invasion inresponse to PRL in all T47D clones (Fig. 5e, Na3VO4).To demonstrate that the role of PAK1 in PRL-mediated

signaling is not limited to T47D cells, we assessed migra-tion and invasion in the presence and absence of Na3VO4in TMX2-28 clones. In TMX2-28 GFP and TMX2-28 WTcells PRL induced cell migration (Fig. 6a, veh) and inva-sion (Fig. 6b, veh) while in TMX2-28 Y3F cells did not(Fig. 6a and b, veh). However, Na3VO4 treatment abol-ished PRL-dependent cell migration and invasion in allTMX2-28 clones (Fig. 6a and b, Na3VO4). Silencing ofPTP-PEST also abolished PRL-dependent cell migrationand invasion of all T47D clones (Fig. 6c and d) and migra-tion of TMX2-28 clones (Fig. 6e). PRL-induced invasionof TMX2-28 control (GFP) and WT cells was significantlydecreased by silencing of PTP-PEST although not com-pletely abolished (Fig. 6f) suggesting that, in addition toPTP-PEST, other tyrosine phosphatases may participate inthe pTyr-PAK1-dependent invasion of TMX2-28 cells.

PAK1 tyrosyl phosphorylation stimulates PRL-inducedtumor metastasis in vivoCell migration is critical for tumor cell metastasis. Inorder to assess whether PRL-induced tyrosyl phosphoryl-ation of PAK1 has a physiological effect on breast cancer

Fig. 4 Silencing of tyrosine phosphatase PTP-PEST rescues FAKauto-phosphorylation in T47D and TMX2-28 cells. a PTP-PEST siRNAreduces PTP-PEST mRNA in T47D and TMX2-28 cells. T47D and TMX2-28 cells were transfected with either PTP-PEST siRNA or controlnon-coding siRNA (ctrl) and mRNA levels were assessed by RT-PCRusing PTP-PEST-specific primers. GAPDH primers were used an internalcontrol. b WCL of T47D and TMX2-28 clones transfected with controlor PTP-PEST siRNAs and treated with PRL (200 ng/ml) for 0 or 15 minwere probed for FAK auto-phosphorylation using αpY397-FAKantibody. The expression levels of γtubulin were used as an internalloading control. c Graph represents the densitometric analysis of thebands obtained for pY397-FAK normalized to total FAK


metastasis, TMX2-28 stably overexpressing GFP, myc-PAK1 WT, or myc-PAK1 Y3F were inoculated in mousemammary fat pads and mice were treated with PRL for8 weeks. Tumors and lungs were harvested and homoge-nized and proteins were separated by SDS-PAGE and ana-lyzed for myc-tagged PAK1 to indicate metastasis of theprimary tumor into distant tissues. We focused on the pri-mary tumor and the lungs, as the lungs are one of themost common sites for secondary tumor in patients withmetastatic breast cancer. As expected, each primary tumor

from all PAK1 WT and PAK1 Y3F mice was positive formyc-tagged PAK1 (Fig. 7a) while GFP cells do not producetumors. Myc-tagged PAK1 was detected in 3 out of 8lungs from the PAK1 WT mice while there was no detect-able myc-PAK1 in any of the PAK1 Y3F or GFP mouselungs (Fig. 7a). Tumors and lungs were also fixed and ana-lyzed by immunocytochemistry (IHC) with anti-myc. OurIHC analysis revealed that anti-myc signal was detected inbreast tumor (B) of myc-PAK1 WT- and myc-PAK1 Y3F-inoculated mice (C) as well as in lung of myc-PAK1 WT-

Fig. 5 Protein tyrosine phosphatase inhibition impedes PRL-mediated T47D cell migration and invasion. a, b Equal amounts of T47D GFP, PAKWT, or PAK1 Y3F cells were loaded into the upper part of the Boyden chamber uncovered (a) or covered with Matrigel (b) with or without Na3VO4(100 ng/ml). PRL (200 ng/ml) was added to the lower part. Representative brightfield images of the cells migrated/invaded to the lower chamberwere taken in 48 h. A LUCPlan FLN 40X objective lens and wide field WHN 10X eyepiece on an inverted Olympus IX81 microscope were used.(c) Na3VO4 (100 ng/ml) treatment on T47D cells for 48 h has no cytotoxic effect. d, e The number of cells that migrated to the lower surface of thechamber toward PRL (white bar) or vehicle (black bar) after 48 h was counted and plotted. Bars represent mean ± SE. *P < 0.05 compared with thesame cells not treated with PRL


inoculated mice (D) but not in lungs of control GFP- (E)or PAK1 Y3F-inoculated mice (F). These data provide firstin vivo evidence that tyrosyl phosphorylation of PAK1plays a significant role in PRL-induced breast cancer cellmotility and metastasis, as only cells overexpressing PAK1WT, but not phospho-tyrosine-deficient PAK1 Y3F, wereable to migrate from the primary tumor to the lungs. Herewe provide new insight into the mechanisms regulatingPRL-dependent breast cancer cell metastasis.

DiscussionThe role of PAK1 in the regulation of cell motility is welldocumented (reviewed in [11]). The role of PAK1 in theregulation of cell adhesion is also well documented and atleast one mechanism has been proposed ([35], reviewed in

[36]). According to this mechanism, PAK1 phosphorylatespaxillin on Ser273, leading to increased paxillin-GIT1binding and adhesion turnover [35]. We have previouslyimplicated PRL/JAK2-dependent tyrosyl phosphorylationof PAK1 in regulation of cell motility and invasion[21, 22]. We have also implicated pTyr-PAK1 in theregulation of breast cancer cell adhesion and demon-strated that phosphorylation of tyrsines 153, 201 and 285of PAK1 regulates cell adhesion, contribute to maximalPAK1 kinase activity and increased ability to bind βPIXand GIT1 [24]. Here we extend our findings and demon-strate that pTyr-PAK1 phosphorylates MEK1 on Ser289resulting in subsequent ERK1/2 activation. We also showthat PRL-induced FAK auto-phosphorylation on Tyr397 isinhibited by pTyr-PAK1 and can be rescued by inhibiting

Fig. 6 Silencing of the tyrosine phosphatase PTP-PEST reduces PRL-mediated cell migration and invasion. a Tyrosine phosphatase inhibition abolishesPRL-induced TMX2-28 cell migration a and invasion b. TMX2-28 GFP, PAK WT, or PAK1 Y3F cells were assessed as in Fig. 5. c–f Equal amount of T47D(c, d) or TMX2-28 (e, f) clones were transfected with either control or PTP-PEST siRNA and loaded into the upper part of the Boyden chamber covered(d, f) or not (c, e) with Matrigel. The number of cells that migrated/invaded to the lower chamber toward PRL (white bar) or vehicle (black bar) after48 h was counted and plotted. Bars represent mean ± SE. *P < 0.05 compared with the same cells not treated with PRL. #P < 0.05 compared with thesame cells treated with PRL but transfected with control siRNA (f)


tyrosine phosphatases and silencing tyrosine phosphatasePTP-PEST. These tyrosine phosphatase inhibitions ab-rogate cell motility and invasion in response to PRL. Wehypothesize that pSer910-FAK recruits tyrosine phosphat-ase PTP-PEST to dephosphorylate pTyr397-FAK andthereby promotes cell motility as shown previously [30].Dynamic of FAK phosphorylation is significant for cell

motility. Previously, FAK activation has been demon-strated to positively regulate cell motility (for review[25–27]) however it is becoming evident that the role ofFAK activation in cell migration is more complex. Silen-cing FAK using siRNA enhanced HeLa cell migration oncollagen, and FAK dephosphorylation on Y397 by thetyrosine phosphatase PTP-PEST promoted Ras-inducedcell migration in transformed NIH3T3-v-H-Ras cells[29, 30]. Cells with reduced FAK dephosphorylation haddiminished cell motility [37] and overexpression of the

tyrosine phosphatase LMR-PTP, which dephosphorylatesFAK, enhanced cell motility [38]. Importantly, Zheng et al.implicated PAK1 in regulation of Ras-induced FAK de-phosphorylation, as overexpression of constitutively activePAK1 T423E promoted FAK dephosphorylation while in-hibition of PAK1 severely abolished FAK dephosphory-lation at Y397 [30]. With agreement with these data, wehave shown here that pTyr-PAK1 abolished PRL-dependent phosphorylation of Ser397-FAK.We previously demonstrated that tyrosyl phosphoryl-

ation of PAK1 promotes both PAK1 kinase activity andprotein-protein interaction capabilities (for review [23]).PAK1 directly binds to ERK in response to adhesion tofibronectin, and both PAK1 and ERK co-localize at nas-cent adhesions on the cell periphery [39]. Here, PAK1 canserve as a scaffold, bringing together Raf, MEK and ERKat cell/matrix adhesions and thereby stimulating ERK-

Fig. 7 PAK1 tyrosyl phosphorylation stimulates PRL-induced tumor metastasis in vivo. a myc-PAK1 was detected in all tumor lysates isolated frommyc-PAK1 WT and myc-PAK1 Y3F inoculated mice. PRL-induced tyrosyl phosphorylation of PAK1increased tumor metastasis, as 3 out of 8 lungsfrom WT mice contained myc-PAK1 (lanes 2, 4 and 5). No myc-PAK1 was detected in any of the lungs from the Y3F or GFP mice. Anti-tubulinantibody was used as a loading control. Whole cell lysate (WCL) of TMX2-28 PAK1 WT cells was loaded as a control for PAK1-myc position in thegels. b–f Representative images of myc-PAK1 detected with anti-myc in breast tumor (b, c) and lung tissue (d–f). myc-PAK1 was detected inbreast tumors of myc-PAK1 WT- (b) and Y3F- (c) and lung of myc-PAK1 WT- inoculated mice (d) but not in lungs of GFP (e) or PAK1 Y3F-inoculatedmice (f). The arrow highlights metastatic nodule. Counterstaining with hematoxylin was omitted. Scale bar is 200 μm in (b) and 100 μm in (c–e)


dependent signal transduction [39]. In addition to PAK1scaffolding activity, PAK1 promotes Raf activation by dir-ectly phosphorylated Raf on S338/339 [40, 41], and stimu-lates MEK/ERK binding and subsequent ERK activity bydirectly phosphorylating S298 on MEK1 [42–44]. Import-antly, pS298-MEK has been shown to localize at periph-eral adhesion complexes in response to cell adhesion tofibronectin [45]. Concurrently, we have demonstrated thattyrosyl phosphorylated PAK1 is localized at peripheral ad-hesion complexes in response to PRL and is responsiblefor proper adhesion turnover, an important process in cellmigration [24]. This is important, as FAK is also localizedat peripheral adhesion complexes and dynamic FAKlocalization and phosphorylation is important for properadhesion turnover and cell migration (for review [25]).FAK localization to peripheral cell/matrix adhesions isdependent on its focal adhesion targeting (FAT) domainand binding to adhesion proteins paxillin and vinculin[46, 47]. Paxillin phosphorylation at Y31 and Y118 byFAK is necessary for cell migration and adhesion turnover[48–50], however, constitutive tyrosyl phosphorylation ofpaxillin impedes cell migration, and dephosphorylation ofFAK by PTP-PEST is required for proper adhesion turn-over in migrating cells [51]. Furthermore, overexpressionof the dominant negative form of protein phosphataseLMR-PTP leads to FAK hyperphosphorylation and re-duced cell motility [38] suggesting that complex regulationof FAK at adhesion complexes is necessary for proper cellmigration. Phosphorylation of the FAK FAT domain onS910 and Y925 by ERK2 and Src, respectively, results in re-duced FAK/paxillin binding and promotes adhesion turn-over [52, 53]. Furthermore, pS298-MEK/ERK activation inNIH3T3 cells was shown to induce FAK dephosphoryla-tion through ERK-mediated FAK S910 phosphorylationand resulting recruitment of tyrosine phosphatase PTP-PEST and thereby promote cell motility [30]. In this regard,PRL-induced tyrosyl phosphorylation of PAK1 and result-ing adhesion localization could be creating localized PAK1/MAPK/FAK signaling at adhesion complexes and promot-ing adhesion turnover during cell migration.In the present study we demonstrated that tyrosyl

phosphorylation of PAK1 stimulates tumor cell metastasisin vivo. These data, combined with an animal studyreporting prevention of neoplasia progression into inva-sive carcinoma in PRL receptor deficient mice [7], suggestthat PRL is involved in the development of metastasis andtumor progression. Thus, our current data on pTyr-PAK1regulation of FAK phosphorylation bring insight into themechanism of PRL-stimulated motility of breast cancercells.

ConclusionsHere we propose a mechanism by which PRL regulatesmotility of T47D and TMX2-28 cells through pTyr-

PAK1, MEK/ERK and FAK that integrates our findingswith previous studies (Fig. 8). In response to PRL, FAKis auto-phosphorylated and PAK1 is tyrosyl phosphory-lated by JAK2, stimulating PAK1 kinase activity and in-creasing PAK1 protein-protein binding abilities. We showthat tyrosyl phosphorylated PAK1 phosphorylates MEK atserine 298, resulting in MEK-mediated ERK1/2 activation.Activated ERK phosphorylates FAK at S910, leading tosubsequent recruitment of PTP-PEST and dephosphoryla-tion of Y397-FAK [30]. PRL-dependent down-regulationof FAK activity may promote focal contact turnoverthereby promoting cell migration. Finally, we demon-strate for the first time that tyrosyl phosphorylation ofPAK1 by PRL increases breast cancer cell metastasis invivo.

AbbreviationsERK, extracellular signal-related kinase; FAK, focal adhesion kinase; JAK2,Janus kinase 2; MEK1, MAPK/ERK kinase 1; PAK1, p21-activated kinase 1;PRL, prolactin; WCL, whole cell lysate

AcknowledgmentsWe thank Dr. Eisenmann (University of Toledo, OH) for providing TMX2-28cells. We thank Dr. Lindner and Ms. Parker (Lerner Research Institute, Cleveland

Fig. 8 Proposed mechanism for the role of PRL-activated PAK1 inbreast cancer cell migration. PRL binding to the PRLR results inactivation of the non-receptor tyrosine kinase JAK2. JAK2 tyrosylphosphorylates PAK1 on Y153, 201, and 285, enhancing PAK1kinase and scaffolding activities. PRL treatment also leads to FAKauto-phosphorylation at Y397. Activated PAK1 phosphorylates MEK1 atS298, resulting in increased MEK1/ERK binding and enhancedERK activity. Active ERK phosphorylates FAK at S910, leading todephosphorylation of FAK at Y397 by the tyrosine phosphatasePTP-PEST as shown by Zheng et al. (2009). FAK dephosphorylationdecreases FAK kinase activity and promotes adhesion turnover andbreast cancer cell migration


Clinic, OH) for in vivo experiments. We also thank Prabesh Khatiwada for helpduring the re-submission of the manuscript.

FundingThis work was supported by a grant from the National Institutes of Health(R01DK88127 to MD).

Availability of data and materialsAll relevant information is included in the manuscript. Please addressrequests for additional data and materials to the corresponding author.

Authors’ contributionsExperiments were designed by AH and MD, and performed by AH.Manuscript was written by AH. Both authors have read and approved thefinal version of the manuscript.

Competing interestsThe authors declare that they have no competing interests.

Consent for publicationNot applicable.

Ethics approval and consent to participateNot applicable.

Received: 3 February 2016 Accepted: 4 August 2016

References1. Marano RJ, Ben-Jonathan N. Minireview: extrapituitary prolactin: an

update on the distribution, regulation, and functions. Mol Endocrinol.2014;28(5):622–33.

2. Clevenger CV, Furth PA, Hankinson SE, Schuler LA. The role of prolactin inmammary carcinoma. Endocr Rev. 2003;24(1):1–27.

3. Bernichtein S, Touraine P, Goffin V. New concepts in prolactin biology.J Endocrinol. 2010;206(1):1–11.

4. Holtkamp W, Nagel GA, Wander HE, Rauschecker HF, von Heyden D.Hyperprolactinemia is an indicator of progressive disease and poorprognosis in advanced breast cancer. Int J Cancer. 1984;34(3):323–8.

5. Bhatavdekar JM, Shah NG, Balar DB, Patel DD, Bhaduri A, Trivedi SN, et al.Plasma prolactin as an indicator of disease progression in advanced breastcancer. Cancer. 1990;65(9):2028–32.

6. Mujagic Z, Mujagic H. Importance of serum prolactin determination inmetastatic breast cancer patients. Croat Med J. 2004;45(2):176–80.

7. Oakes SR, Robertson FG, Kench JG, Gardiner-Garden M, Wand MP, Green JE,et al. Loss of mammary epithelial prolactin receptor delays tumor formationby reducing cell proliferation in low-grade preinvasive lesions. Oncogene.2007;26(4):543–53.

8. Maus MV, Reilly SC, Clevenger CV. Prolactin as a chemoattractant for humanbreast carcinoma. Endocrinology. 1999;140(11):5447–50.

9. Liby K, Neltner B, Mohamet L, Menchen L, Ben-Jonathan N. Prolactinoverexpression by MDA-MB-435 human breast cancer cells acceleratestumor growth. Breast Cancer Res Treat. 2003;79(2):241–52.

10. Rider L, Shatrova A, Feener EP, Webb L, Diakonova M. JAK2 tyrosine kinasephosphorylates PAK1 and regulates PAK1 activity and functions. J BiolChem. 2007;282(42):30985–96.

11. Molli PR, Li DQ, Murray BW, Rayala SK, Kumar R. PAK signaling inoncogenesis. Oncogene. 2009;28(28):2545–55.

12. Kumar R, Gururaj AE, Barnes CJ. p21-activated kinases in cancer. Nat RevCancer. 2006;6(6):459–71.

13. Bekri S, Adelaide J, Merscher S, Grosgeorge J, Caroli-Bosc F, Perucca-Lostanlen D, et al. Detailed map of a region commonly amplifiedat 11q13→q14 in human breast carcinoma. Cytogenet Cell Genet.1997;79(1–2):125–31.

14. Ong CC, Jubb AM, Haverty PM, Zhou W, Tran V, Truong T, et al. Targetingp21-activated kinase 1 (PAK1) to induce apoptosis of tumor cells. Proc NatlAcad Sci U S A. 2011;108(17):7177–82.

15. Balasenthil S, Sahin AA, Barnes CJ, Wang RA, Pestell RG, Vadlamudi RK, et al.p21-activated kinase-1 signaling mediates cyclin D1 expression in mammaryepithelial and cancer cells. J Biol Chem. 2004;279(2):1422–8.

16. Salh B, Marotta A, Wagey R, Sayed M, Pelech S. Dysregulation ofphosphatidylinositol 3-kinase and downstream effectors in human breastcancer. Int J Cancer. 2002;98(1):148–54.

17. Vadlamudi RK, Adam L, Wang RA, Mandal M, Nguyen D, Sahin A, et al.Regulatable expression of p21-activated kinase-1 promotes anchorage-independent growth and abnormal organization of mitotic spindles inhuman epithelial breast cancer cells. J Biol Chem. 2000;275(46):36238–44.

18. Holm C, Rayala S, Jirstrom K, Stal O, Kumar R, Landberg G. Associationbetween Pak1 expression and subcellular localization and tamoxifenresistance in breast cancer patients. J Natl Cancer Inst. 2006;98(10):671–80.

19. Wang RA, Zhang H, Balasenthil S, Medina D, Kumar R. PAK1hyperactivation is sufficient for mammary gland tumor formation.Oncogene. 2006;25(20):2931–6.

20. Adam L, Vadlamudi R, Mandal M, Chernoff J, Kumar R. Regulation ofmicrofilament reorganization and invasiveness of breast cancer cells bykinase dead p21-activated kinase-1. J Biol Chem. 2000;275(16):12041–50.

21. Rider L, Oladimeji P, Diakonova M. PAK1 regulates breast cancer cell invasionthrough secretion of matrix metalloproteinases in response to prolactin andthree-dimensional collagen IV. Mol Endocrinol. 2013;27(7):1048–64.

22. Hammer A, Rider L, Oladimeji P, Cook L, Li Q, Mattingly RR, et al. Tyrosylphosphorylated PAK1 regulates breast cancer cell motility in response toprolactin through filamin a. Mol Endocrinol. 2013;27(3):455–65.

23. Hammer A, Diakonova M. Tyrosyl phosphorylated serine-threonine kinasePAK1 is a novel regulator of prolactin-dependent breast cancer cell motilityand invasion. Adv Exp Med Biol. 2015;846:97–137.

24. Hammer A, Oladimeji P, De Las Casas LE, Diakonova M. Phosphorylation oftyrosine 285 of PAK1 facilitates betaPIX/GIT1 binding and adhesionturnover. FASEB J. 2015;29(3):943–59.

25. Mitra SK, Hanson DA, Schlaepfer DD. Focal adhesion kinase: in commandand control of cell motility. Nat Rev Mol Cell Biol. 2005;6(1):56–68.

26. Parsons JT. Focal adhesion kinase: the first 10 years. J Cell Sci.2003;116(Pt 8):1409–16.

27. Hanks SK, Ryzhova L, Shin NY, Brabek J. Focal adhesion kinase signalingactivities and their implications in the control of cell survival and motility.Front Biosci. 2003;8:d982–96.

28. Schaller MD. FAK and paxillin: regulators of N-cadherin adhesion andinhibitors of cell migration? J Cell Biol. 2004;166(2):157–9.

29. Yano H, Mazaki Y, Kurokawa K, Hanks SK, Matsuda M, Sabe H. Roles playedby a subset of integrin signaling molecules in cadherin-based cell-celladhesion. J Cell Biol. 2004;166(2):283–95.

30. Zheng Y, Xia Y, Hawke D, Halle M, Tremblay ML, Gao X, et al. FAKphosphorylation by ERK primes ras-induced tyrosine dephosphorylation ofFAK mediated by PIN1 and PTP-PEST. Mol Cell. 2009;35(1):11–25.

31. Fasco MJ, Amin A, Pentecost BT, Yang Y, Gierthy JF. Phenotypic changes inMCF-7 cells during prolonged exposure to tamoxifen. Mol Cell Endocrinol.2003;206(1–2):33–47.

32. Schlaepfer DD, Mitra SK, Ilic D. Control of motile and invasive cell phenotypesby focal adhesion kinase. Biochim Biophys Acta. 2004;1692(2–3):77–102.

33. Payne DM, Rossomando AJ, Martino P, Erickson AK, Her JH, Shabanowitz J,et al. Identification of the regulatory phosphorylation sites in pp 42/mitogen-activated protein kinase (MAP kinase). EMBO J. 1991;10(4):885–92.

34. Zhang J, Zhang F, Ebert D, Cobb MH, Goldsmith EJ. Activity of theMAP kinase ERK2 is controlled by a flexible surface loop. Structure.1995;3(3):299–307.

35. Nayal A, Webb DJ, Brown CM, Schaefer EM, Vicente-Manzanares M, HorwitzAR. Paxillin phosphorylation at Ser273 localizes a GIT1-PIX-PAK complex andregulates adhesion and protrusion dynamics. J Cell Biol. 2006;173(4):587–9.

36. Parrini MC. Untangling the complexity of PAK1 dynamics: the futurechallenge. Cell Logist. 2012;2(2):78–83.

37. Yu DH, Qu CK, Henegariu O, Lu X, Feng GS. Protein-tyrosine phosphataseShp-2 regulates cell spreading, migration, and focal adhesion. J Biol Chem.1998;273(33):21125–31.

38. Rigacci S, Rovida E, Dello Sbarba P, Berti A. Low Mr phosphotyrosine proteinphosphatase associates and dephosphorylates p125 focal adhesion kinase,interfering with cell motility and spreading. J Biol Chem. 2002;277(44):41631–6.

39. Sundberg-Smith LJ, Doherty JT, Mack CP, Taylor JM. Adhesion stimulatesdirect PAK1/ERK2 association and leads to ERK-dependent PAK1 Thr212phosphorylation. J Biol Chem. 2005;280(3):2055–64.

40. Chaudhary A, King WG, Mattaliano MD, Frost JA, Diaz B, Morrison DK, et al.Phosphatidylinositol 3-kinase regulates Raf1 through Pak phosphorylation ofserine 338. Curr Biol. 2000;10(9):551–4.


41. Zang M, Hayne C, Luo Z. Interaction between active Pak1 and Raf-1 isnecessary for phosphorylation and activation of Raf-1. J Biol Chem.2002;277(6):4395–405.

42. Frost JA, Steen H, Shapiro P, Lewis T, Ahn N, Shaw PE, et al. Cross-cascadeactivation of ERKs and ternary complex factors by Rho family proteins.Embo J. 1997;16(21):6426–38.

43. Coles LC, Shaw PE. PAK1 primes MEK1 for phosphorylation by Raf-1kinase during cross-cascade activation of the ERK pathway. Oncogene.2002;21(14):2236–44.

44. Park ER, Eblen ST, Catling AD. MEK1 activation by PAK: a novel mechanism.Cell Signal. 2007;19(7):1488–96.

45. Slack-Davis JK, Eblen ST, Zecevic M, Boerner SA, Tarcsafalvi A, Diaz HB, et al.PAK1 phosphorylation of MEK1 regulates fibronectin-stimulated MAPKactivation. J Cell Biol. 2003;162(2):281–91.

46. Tachibana K, Sato T, D’Avirro N, Morimoto C. Direct association of pp125FAKwith paxillin, the focal adhesion-targeting mechanism of pp125FAK. J ExpMed. 1995;182(4):1089–99.

47. Chen HC, Appeddu PA, Parsons JT, Hildebrand JD, Schaller MD, Guan JL.Interaction of focal adhesion kinase with cytoskeletal protein talin. J BiolChem. 1995;270(28):16995–9.

48. Petit V, Boyer B, Lentz D, Turner CE, Thiery JP, Valles AM. Phosphorylation oftyrosine residues 31 and 118 on paxillin regulates cell migration through anassociation with CRK in NBT-II cells. J Cell Biol. 2000;148(5):957–70.

49. Webb DJ, Donais K, Whitmore LA, Thomas SM, Turner CE, Parsons JT, et al.FAK-Src signalling through paxillin, ERK and MLCK regulates adhesiondisassembly. Nat Cell Biol. 2004;6(2):154–61.

50. Zaidel-Bar R, Milo R, Kam Z, Geiger B. A paxillin tyrosine phosphorylationswitch regulates the assembly and form of cell-matrix adhesions. J Cell Sci.2007;120(Pt 1):137–48.

51. Angers-Loustau A, Cote JF, Charest A, Dowbenko D, Spencer S, Lasky LA, et al.Protein tyrosine phosphatase-PEST regulates focal adhesion disassembly,migration, and cytokinesis in fibroblasts. J Cell Biol. 1999;144(5):1019–31.

52. Hunger-Glaser I, Fan RS, Perez-Salazar E, Rozengurt E. PDGF and FGF inducefocal adhesion kinase (FAK) phosphorylation at Ser-910: dissociation fromTyr-397 phosphorylation and requirement for ERK activation. J Cell Physiol.2004;200(2):213–22.

53. Katz BZ, Romer L, Miyamoto S, Volberg T, Matsumoto K, Cukierman E, et al.Targeting membrane-localized focal adhesion kinase to focal adhesions:roles of tyrosine phosphorylation and SRC family kinases. J Biol Chem.2003;278(31):29115–20.

• We accept pre-submission inquiries • Our selector tool helps you to find the most relevant journal• We provide round the clock customer support • Convenient online submission• Thorough peer review• Inclusion in PubMed and all major indexing services • Maximum visibility for your research

Submit your manuscript atwww.biomedcentral.com/submit

Submit your next manuscript to BioMed Central and we will help you at every step:


AbstractBackgroundResultsConclusion

BackgroundMethodsAntibodies and reagentsCell cultureAssessing FAK, MEK, and ERK phosphorylationPTP-PEST knockdownCell viabilityCell migration and cell invasion assaysIn vivo metastasisStatistical analysis

ResultsTyrosyl phosphorylated PAK1 negatively regulates FAK auto-phosphorylationTyrosyl phosphorylation of PAK1 promotes S298-MEK1 phosphorylation and ERK activation in response to PRLProtein tyrosine phosphatase inhibition rescues �PRL-mediated auto-phosphorylation of FAKProtein tyrosine phosphatase inhibition impedes PRL-mediated T47D and TMX2-28 cell migration and invasionPAK1 tyrosyl phosphorylation stimulates PRL-induced tumor metastasis in vivo

DiscussionConclusionsAbbreviationsAcknowledgmentsFundingAvailability of data and materialsAuthors’ contributionsCompeting interestsConsent for publicationEthics approval and consent to participateReferences

prolactin-induced pak1 tyrosyl phosphorylation promotes ...[17], while expression of kinase dead pak...

Documents